Inorg. Chem. 1991, 30, 3361-3362
3361
Table 11. Observed and Calculated Electronic Spectral Data for [Co(N,)(acac)]X Complexes abs bands
solution‘ comDd [Co(3,3,3-N4)(acac)]PF6
[Co(3,2,3-N4)(acac)]PF6
[co(2,3,2-N4)(acac)]PF6
A. nm 1009 49 1
8 34
9910 20 370
292 960 476
10800 9
979 458
10 220 21 830
820
1 I49
10543 21 320 22 033
as above
51
34 250 10 420 21 010
292 995 485
1 lo00 9 50
34 250 10 100 20 620
982 458
10 180 21 830
10232
as above
795
1113
20 670
289 1020 49 1
7000 8 31
34 600 9 800 20 370
1026 488
9 750 20 490
1080
10 374 20 072 21 174
as above
772
293 999 486
9100 8 37
34 130 10010 20 580
983 468
10 170 21 370
1103
10 144 20488 21 174
as above
788
292 990
34 250 10 100 20 750
990 485
10 100 20 620
795
1113
10232 20 670 21 362
as above
482
10300 6 40
292
11200
34 250
Y.
cm-I
solid stateb Y, cm-l
X. nm 985 476
tC
8’.cm-’
Du,cm-’
calcd Y, cm-’
10 150 21 010
780
1092
10039 20 280 20 959
assprnts
21 362
[Co(3,3,3-N4)(acac)]l
[Co(3,2,3-N,)(acac)]I
[Co(2,3,2-N4)(acac)]I
M (methanol), and -1.0 X DSolutionspectra were obtained by using concentrations of -1 X lo-’ M (tetrachloroethylene, C2CI,), -1.0 X lo-’ M (methanol), one for each band going from low energy to high energy. bSolid-state spectra were obtained by using the Nujol mull diffuse-
transmittance techniq~e.~’CMolarabsorptivities, L mol-’ cm-’.
Both visual and numerical comparison of the diffraction lines show that the patterns are identical and that the compounds are isomorphic. Conclusions While there is no doubt that the 12 complexes prepared in this study have pseudooctahedral structures, visible spectra cannot be used to determine whether the compounds exist in the a-cis or 0 4 s isomeric forms or a mixture thereof. Proton magnetic resonance studies are precluded by the paramagnetic nature of both metal ions. This leaves infrared spectral data as the only source for conjecture. In the infrared spectra of [M(2,3,2-N4)(acac)]PF6 and [M(3,2,3-N4)(acac)]PFs, the presence of only one isomer is suggested by the uncomplicated nature of the nitrogen-hydrogen stretching vibrations observed. A mixture of isomers may be indicated by the more complicated patterns of the corresponding iodide salts; however, in these compounds, selective hydrogen bonding of one set of amine protons to the iodide ion causes a shift of some of these stretching vibrations to lower energies. Infrared spectra of the [M( 3,3,3-N4)(acac)] PF6 and [M( 3,3,3-N,)(acac)] I complexes show the most complicated stretching patterns for the amine protons and are most likely to exist as a mixture of isomers because of greater flexibility associated with longer carbon chains. Indeed, isomer studies of tetraamine complexes32have shown that as the chain length increases, the trans form is favored and only the &cis form exists in the presence of a bidentate ligand. It is likely that the &cis form dominates in the (3,2,3) and (3,3,3) complexes prepared in this study, but the differences in numbers of IR bands may be due to solid-state effects only. Attempts to prepare macrocyclic tetraaza complexes via the acid-catalyzed (pH 3) rearrangement of the [Ni(N,)(acac)]X complexes in aqueous solution were unsuccessful. Three hours of refluxing followed by addition of 1 equiv of KX always resulted in the formation of the corresponding tetraamine, [Ni(N,)X,], which was identified on the basis of its infrared spectrum. Under the same conditions, nickel complexes containing similar sexadentate Schiff base ligands do react to form the desired macrocycle~.~J’This suggests that 2 equiv of &diketone are necessary
for ring closure to occur. This hypothesis was tested by refluxing acidic solutions of the [Ni(N4)(acac)]X complexes in the presence of an additional 1 equiv of 2,4-pentanedione. Small amounts of the macrocyclic species are produced;33 however, the mechanism involved here cannot be distinguished from the in situ procedure previously r e p ~ r t e d . ~The . ~ latter procedure continues to provide the best means of synthesis of macrocyclic complexes with uninegative Schiff base ligands. Acknowledgment. S.C.C. gratefully acknowledges financial support of this investigation by the National Institutes of Health, Grant HL 15640 from the Heart, Lung, and Blood Institute. Supplementary Material Available: Tables of analytical data along with yields and colors of individual complexes and magnetic susceptibility data (1 page). Ordering information is given on any current masthead page.
(32) Alexander, M.D.; Hamilton, G . Inorg. Chem. 1969, 10, 2131.
Chemistry, University of Florida, Gainesville, FL 32611.
-
0020-1 669/91/ 1330-3361$02.50/0
(33) Sweitzer, R.K.; Cumming, S . C. Proceedings ofthe Undergraduate Science Symposium; Vol. I, Central State University: Wilberforce. OH 45384, 1984; Vol. I, p 26.
Contribution from the Department of Chemistry, The University of Texas at Austin, Austin, Texas 78712 A Theoretical Study of the HP, Ion Michael J. S.Dewar* and Ya-Jun Zheng Received March 8, 1990
Baudler et a1.l recently reported the reduction of tetraphosphorus (1) to an anion (2 or 3) which can be regarded as the monoconjugate base from tetraphosphabicyclobutane(4). The NMR spectrum of the ion in solution showed three different ‘To whom correspondence should bc addressed at the Department of
0 1991 American Chemical Society
3362 Inorganic Chemistry, Vol. 30, No. 17. 1991
Notes Tible 1. Calculated Properties of P, Compounds
bond Ienz?ths. - .A
I
2
3
compd 1 2 3 4 7
AH?
PI-P2
P2-P3
P2-P4
50.4 -13.4 -25.0 36.2 -22.8
2.040
2.040 1.974 2.037 2.048 2.015
2.040
2.019 1.960 2.014 2.W9
2.105 2.022 2.014 2.007
#Heat of formation (kcal/mol). 4
5
6
Table It. Calculated Formal Atomic Charges
formal atomic charges. at camad 2 3 4 1
7
- -
phosphorus resonanas, implying that degenerate rearrangement by migration of hydrogen from P4 to PI (i.e. 2 5 or 3 6) is slow on the NMR time scale. This suggests that the ion has the anti structure (2) because interconversion of the syn isomers (3and 6) would be expected to lake place very rapidly, given that it involves only a small displacement of hydrogen and given that it could take place by tunneling. Furthermore, other analogous phosphorus compounds have been shown*to have anti geometries. Since AMI3 parameters are now available' for phosphorus, we have used them to studv 1-4. The calculations were carried out by using the standard-AM1 procedure as implemented in the AMPAC2.1 program? The calculated geometries and heats of formation arc shown in Table I, and the calculated formal charges, in Table 11. A M 1 seemingly overestimates strain energies due to P-P-P bending. the heat of formation calculated for I being too positive by 36.3 kcal/mol? The hcau of formation calculated for 1-8 are therefore probably all too large, by a lesxr amount. However. since the strain energies of all these species should in any case be similar, the errors should not affect their relative enerpies. which are the quantities of interest here. The syn isomer 3 is predicted to be not only lower in energy than 2 but much lower, the difference between their calculated heats of formation being 11.5 kcal/mol. While the AMI parameters for phosphorus have not been tested as thoroughly as those for the 'organic" elements. these results seem to suggest strongly that 3 is indeed the more stable isomer. This would eliminate the most natural explanation for the relative slowness of the interamversion of 3 and 6. We also examined the alternative possibility. that interconversion of 3 and 6 might be slow on the NMR time scale. Houcver, the calculated barrier was, as expected, very low (2.2 kcal/mol). The properties of the corresponding transition state (7). which had C, symmetry. are shown in Tables I and II. The results in Table I also indicate that 4 is a strong acid, its calculated deprotonation energy being only 306 kcal/mol. AMI has been shown6 to give good ertimatcs of proton affinities and deprotonation energies. The exwrimental studies were carried out for the sodium salt of 2 (3) in solution whereas our calculations refer to the isolated anions. The apparent discrepancy between them might then be due to ion pairing. Ion pairing might either selectively stabilize the anti isomer (2) or inhibit the rearrangement of 3. (I) Baudlcr. M.; Adamck, C.: Opich. S.: Budzikiruia, H.:Oumnis, D.
Angcw. Chrm., Inl. Ed. Engl. 1988, 27, 1059. (2) Palcnik. 0. J.: Donohuc, J. Acta C?ystallopr. 1962. IS, 564. (3) Dewar. M.J.S.:Zabiach. E.O.:Healy. E.F.:Stmarl. J. J. P.J.Am. Chrm. Soe. 1985, 107, 3902. (4) Dewar, M.J. S.: lie, C. THEOCHEM. in pnss. (5) Available from the Quantum Chemistry Program Exchange (QCPE): Program 506. (6) Dewar. M.J. S.;Dieter, K. M. J. Am. Chem. Soc. 1%. 108,8075.
PI
PZ
P4
H
-0.6430 -0.8670 0.1220 4.7727
-0.0130 -0.0995 4,1751 0.1098
-0.4010 -0.5212 0.1219 -0.7804
0.0701 0.1898 0.0531 0.3332
.In units of the electronic charge. H
..^
1.3752
-0
Flyre 1. and 2.
Geometry calculated f a the adduct bctween a sparkle (M*)
Since AM1 parameters are not yet available for sodium or potassium, we checked these possibilities by using a disembodied unit positive charge ('spa~kle")~as a model for the melal cation, placing it near the negatively charged phosphorus atom in 2 or 3. The resulting "ion pair" (8) still had syn hydrogen, its structure being shown in Figure 1. However, conversion of 8 to the corresponding species derived from 6, without moving the sparkle, was stongly endothermic (AH 13 kcal/mol) and no path of lower energy was found for the hydrogen migration. Our results thus support the second suggestion above, i.e. that 3 rearranges slowly in solution because of inhibition by ion pairing. The situation is reminiscent of that involved in the reduction of ketones by sodium borohydride in aprotic solvents, where association of Na+ with the carbonyl oxygen facilitates reaction:'^'
+
H,BH- C=O.-Na+
-
H,B H-C--O-Na+
(I)
If the sodium ions are prevented from associating with carbonyl by adding crown-6 ether, the reaction is completely inhibited8 If the rearrangement of 3 is i n d d inhibited by ion pairing, it should then take place rapidly in the presence of a crown ether able to sequestrate the metal ion. Afknowledgment. This work was supported by grants from the Air Force Office of Scientific Research (AFOSR 89-0179). the Welch Foundation (F-126). and the National Science Foundation (CHE87-12022). Reghistry No. HP;,
134682-45-4.
(7) Dewar, M. J. S.;McKa. M.L.J. Am. Chrm. Soe. 1978.IW. 7499. (8) (a) Pi-. J. L.; Handel. H.TefmhedronMt. 1974.2317. (b) Pierre. I. L.: Handel, H.: Perraud. R. Tefrahedmn 1975.31.2795. (c) Handel. H.:Pierre. J. L. Telrahcdmn 1975. 31. 997. (d) Handel. H.:Pierre. J. L. Telrohcdmn Lett. 1976,2029.
